CAPS National Sexual Health Survey (NSHS)

Abstract

The CAPS National Sexual Health Survey (NSHS) represents a landmark epidemiological and psychometric surveillance instrument developed by the Center for AIDS Prevention Studies (CAPS) at the University of California, San Francisco. Originally formulated to elucidate the behavioral, cognitive, affective, and structural determinants of human immunodeficiency virus (HIV) transmission and sexually transmitted infections (STIs) across the United States, the survey establishes rigorous population-level assessments of sexual health practices. Administered via probability-based random-digit dialing (RDD) telephone sampling among non-institutionalized adults aged 18 years and older across the 48 contiguous states, the NSHS incorporates an intricate multi-tier questionnaire architecture. The instrument spans core epidemiological screeners, detailed dyadic sexual encounter inventories tracking up to ten individual partners in the preceding 12 months, barrier prophylaxis trajectories (including condom slips, breaks, and attitudes), HIV testing behaviors, perceived vulnerability and optimistic bias indices, sexual dysfunction profiles, and validated psychological personality scales assessing sensation-seeking and gender-role adherence constructs such as machismo. The screening and respondent-selection mechanism operationalizes an unclustered next-birthday selection algorithm, standardized callback scheduling protocols, and structured verbal informed consent frameworks. Psychometrically, the instrument demonstrates robust content and construct validity, strong criterion-related predictive validity regarding STI acquisition and high-risk sexual practices, and high test-retest reliability across its structured behavioral matrices. Factor analytic examinations of its embedded latent psychometric scales confirm well-delineated dimensions of sexual risk-taking, barrier negotiation self-efficacy, and affective barrier appraisals. This comprehensive instrument continues to provide a foundational psychometric and methodological blueprint for contemporary surveillance systems, sexual health intervention trials, and behavioral public health research.

Keywords

National Sexual Health Survey, CAPS, HIV risk behavior, condom attitudes, sexual epidemiology, random digit dialing, sensation seeking, optimistic bias, sexual partner enumeration, psychometrics

Authors

The CAPS National Sexual Health Survey was designed, operationalized, and validated by an interdisciplinary consortium of behavioral scientists, biostatisticians, epidemiologists, and survey methodologists affiliated with the Center for AIDS Prevention Studies (CAPS), located within the School of Medicine and Department of Epidemiology and Biostatistics at the University of California, San Francisco (UCSF).

• Joseph A. Catania, Ph.D. — Principal Investigator; primary architect of the AIDS Behavioral Research Project and Professor at the Center for AIDS Prevention Studies, Department of Medicine, University of California, San Francisco. Specialization: Health psychology, sexual risk assessment, dyadic sexual behavior, and survey methodologies in stigmatized health domains.
• Thomas J. Coates, Ph.D. — Co-Principal Investigator; Director of the Center for AIDS Prevention Studies (CAPS) and Professor of Medicine and Epidemiology at UCSF. Specialization: Behavioral medicine, community-level public health interventions, and international HIV/AIDS prevention paradigms.
• Ron Stall, Ph.D., M.P.H. — Co-Investigator; Associate Professor at CAPS/UCSF. Specialization: Medical anthropology, substance use epidemiology, syndemic theory, and urban health disparities.
• Heather A. Turner, Ph.D. — Research Associate and Sociologist; University of California, San Francisco. Specialization: Stress processes, interpersonal victimization, and life-course developmental sociology.
• John L. Peterson, Ph.D. — Research Psychologist; CAPS/UCSF. Specialization: Multicultural health disparities, minority mental health, and psychosocial determinants of sexual health.
• Norman Hearst, M.D., M.P.H. — Associate Professor of Family and Community Medicine and Epidemiology/Biostatistics; CAPS/UCSF. Specialization: International epidemiology, partner notification paradigms, and primary prevention strategies.
• M. Margaret Dolcini, Ph.D. — Developmental and Health Psychologist; CAPS/UCSF. Specialization: Adolescent and young adult sexuality, psychometric scale development, and behavioral change processes.
• Estie Hudes, Ph.D., M.P.H. — Senior Biostatistician; CAPS/UCSF. Specialization: Latent variable modeling, longitudinal cohort analytics, and complex survey variance estimation.
• John H. Gagnon, Ph.D. — Senior Scientific Consultant; Professor of Sociology, State University of New York (SUNY) at Stony Brook. Specialization: Sexual scripting theory, social organization of sexuality, and human sexual conduct.
• James A. Wiley, Ph.D. — Senior Survey Methodologist and Sociologist; Survey Research Center, University of California, Berkeley. Specialization: Probability sampling design, mathematical modeling of epidemic transmission, and survey error modeling.
• Robert M. Groves, Ph.D. — Survey Methodologist; Survey Research Center, Institute for Social Research, University of Michigan. Specialization: Total survey error frameworks, nonresponse bias analytics, and telephone survey administration protocols.

Purpose

The primary purpose of the CAPS National Sexual Health Survey (NSHS) was to address a critical national epidemiological data deficit during the emergent phase of the HIV/AIDS epidemic in the United States. Prior to its deployment, national population-level surveillance regarding human sexual behaviors was predominantly constrained to fertility-focused instruments, such as the National Survey of Family Growth (NSFG), or relied upon non-representative, convenience-based samples that could not generalize to the broader domestic populace. The NSHS was explicitly engineered to obtain unbiased, nationally representative prevalence estimates of HIV-related risk factors, barrier contraceptive behaviors, and systemic sexual health profiles among heterosexual, bisexual, and homosexual populations residing in the 48 contiguous United States.

Beyond broad epidemiological quantification, the theoretical purpose of the survey was to rigorously isolate the psychosocial, cognitive, relational, and demographic mechanisms governing barrier prophylaxis adoption. The instrument was developed to elucidate why individuals exposed to public health messaging frequently failed to consistently utilize barrier protection, how cognitive distortions such as optimistic bias modulated vulnerability perception, and in what manner interpersonal power dynamics influenced sexual safety negotiation. In doing so, the NSHS bridged the divide between structural epidemiological modeling and individual-level psychological inquiry.

In clinical and public health application, the survey yields granular behavioral inventories designed to inform federal, state, and community-level AIDS prevention programming. The data generate essential baseline metrics regarding the distribution of multi-partner sexual networks, the mechanics of barrier failure (differentiating between user error, condom slippage, and manufacturing breakage), and patterns of diagnostic testing engagement. Methodologically, the screener section of the NSHS resolves household enumeration challenges through standardized probabilistic selection (the next-birthday method), safeguarding the scientific integrity of downstream epidemiological inferences.

Psychological Construct

The NSHS operates as a comprehensive battery capturing multidimensional psychological and behavioral constructs. Rather than viewing sexual behavior strictly through an epidemiological lens of mechanical exposure, the instrument operationalizes human sexuality as a multifaceted psychological construct embedded in cognitive appraisal, personality disposition, interpersonal negotiation, and gendered social conditioning.

Perceived Risk and Optimistic Bias

The construct of perceived risk within the NSHS captures both absolute and comparative subjective estimates of acquiring HIV and other STIs. Grounded in cognitive psychological paradigms, the survey measures optimistic bias—the systematic cognitive distortion wherein individuals judge their personal susceptibility to adverse health outcomes as significantly lower than that of their demographic peers, despite engaging in identical risk behaviors. The scale probes personal vulnerability across distinct operational frames: likelihood ratings, comparative self-versus-other assessments, and emotional anxiety regarding contagion.

Condom Attitudes and Affective Prophylaxis Appraisals

Barrier utilization is measured through a latent construct reflecting attitudes toward condoms. This domain is segmented into:

• Affective and Sensory Interference: Appraisals regarding whether condoms diminish tactile pleasure, interrupt emotional intimacy, or induce physiological erectile disruption.
• Interpersonal Trust and Social Stigma: Beliefs that suggesting barrier use conveys infidelity, disease suspicion, or interpersonal mistrust to a romantic partner.
• Efficacy and Safety Confidence: Cognitive trust in the mechanical integrity of condoms to avert pathogenic transmission and unintended pregnancy.

Sensation-Seeking Disposition

Rooted in trait psychometrics, the embedded sensation-seeking scale assesses the biological and psychological disposition toward novel, intense, and varied sensory stimulation, coupled with the willingness to undertake physical, social, and legal risks for the sake of such experiences. In the context of sexual health, this construct assesses spontaneous sexual decision-making, the pursuit of non-monogamous encounters, and the psychological tolerance for sexual disinhibition under conditions of environmental arousal.

Machismo and Gender-Role Hypermasculinity

The survey incorporates culturally responsive psychological scales assessing gender-role ideology, specifically focusing on machismo and traditional male role norms. This construct examines the endorsement of male emotional stoicism, sexual dominance, compulsory heteronormative virility, and resistant attitudes toward female reproductive and barrier negotiation. High scores reflect entrenched normative beliefs that direct prophylaxis negotiation diminishes masculine authority.

Dyadic Sexual Architecture and Partner Characteristics

The behavioral measurement framework evaluates sexual activity across up to ten distinct sexual partnerships occurring within the antecedent 12-month period. For each partnership, the construct captures relational classification (primary, casual, commercial), sexual practice repertoires (vaginal, oral, anal intercourse), exact counts of protected versus unprotected events, history of concurrent substance use during intercourse, and perceived partner risk profiles (known intravenous drug use, partner concurrent relationships, known bisexual history).

Theoretical Framework

The architecture of the CAPS National Sexual Health Survey is predominantly grounded in the AIDS Risk Reduction Model (ARRM), conceptualized by Catania, Coates, and Kegeles in 1990. ARRM is a triphasic, dynamic health psychology framework that integrates constructs from the Health Belief Model, the Theory of Reasoned Action, social cognitive theory, and self-regulation paradigms to model the psychological trajectory of sexual behavior change.

Stage 1: Recognition and Labeling of Sexual Risk

The initial phase posits that individuals cannot and will not modify their sexual behaviors until they accurately perceive that their practices place them at risk for HIV/STIs. The theoretical framework conceptualizes this stage through cognitive appraisal: the acquisition of basic viral transmission knowledge, personal vulnerability estimation, and overcoming optimistic bias. The NSHS operationalizes this stage via items assessing perceived self-vulnerability, general transmission knowledge, and history of diagnostic testing.

Stage 2: Commitment to Behavioral Change

Once risk is cognitively acknowledged, an individual evaluates whether to make an explicit commitment to reduce sexual risk (e.g., initiating consistent condom use, reducing partner numbers, abstaining from penetrative intercourse). This decision is theoretically dictated by a decisional balance matrix—weighing the perceived costs against the perceived benefits of the risk-reduction strategy. In the NSHS, this theoretical mechanism is assessed via condom outcome expectancies, peer social norms, perceived social support, and affective evaluation of barrier methods.

Stage 3: Enactment and Action

The final phase addresses the actual performance of the health-protective behavior. ARRM emphasizes that behavioral enactment is contingent upon three critical operational factors: self-efficacy, interpersonal communication skills, and environmental facilitation. The NSHS captures this domain through partner-specific behavioral negotiation items, history of condom failure handling (responses to slips and breaks), and the ability to maintain behavioral integrity in emotionally charged or chemically intoxicated environments.

Integration of Scripting and Personality Theories

Additionally, the NSHS integrates Sexual Scripting Theory (Gagnon & Simon), which conceptualizes sexual acts as structured psychological scripts dictated by cultural scenarios, interpersonal expectations, and intrapsychic fantasies. By integrating standardized scales measuring sensation seeking (Zuckerman) and traditional gender ideologies (machismo), the NSHS framework acknowledges that cognitive deliberation operates within personality traits and sociocultural expectations regarding masculine and feminine sexual conduct.

Validity

The psychometric validity of the CAPS National Sexual Health Survey has been rigorously evaluated across epidemiological cohorts, utilizing diverse statistical methodologies to confirm that the instrument accurately captures its designated behavioral and psychological constructs.

Content Validity

Content validity was established through an extensive developmental phase overseen by an interdisciplinary panel of psychometricians, epidemiologists, medical anthropologists, and clinical sexologists at UCSF and the Survey Research Center at UC Berkeley. Every prospective behavioral item underwent cognitive debriefing and pilot laboratory pre-testing to verify semantic clarity, minimize respondent cognitive burden, and ensure cultural appropriateness across diverse socio-demographic strata. The respondent selection screener (utilizing the next-birthday protocol) was implemented to eliminate interviewer bias and within-household systematic selection distortion, providing strong methodological content validity.

Construct and Structural Validity

Construct validity for the embedded psychological scales has been corroborated through structural equation modeling (SEM) and confirmatory factor analysis (CFA). The latent dimensions postulated by the AIDS Risk Reduction Model—namely, perceived risk, outcome expectancies, and self-efficacy—demonstrated substantial factor saturation, with standardized factor loadings consistently exceeding .60 across subscales. For the sensation-seeking inventory, items assessing sexual adventure, physical novelty, and psychological disinhibition loaded onto distinct, coherent latent factors that separated cleanly from general anxiety or psychological distress indices.

Convergent and Discriminant Validity

Convergent validity has been repeatedly demonstrated across clinical and survey datasets:

• Scores on the NSHS Condom Attitude Scale correlate substantially with measured behavioral compliance, yielding strong positive associations with lifetime condom utilization rates ($r = .52, p < .001$) and prospective longitudinal condom consistency ($r = .44, p < .001$).
• Measures of perceived risk demonstrate convergent alignment with self-reported epidemiological partner history, wherein individuals reporting higher numbers of concurrent partners or anonymous encounters score significantly higher on the vulnerability continuum ($F > 45.0, p < .001$).
• Discriminant validity is supported by weak or non-significant correlations between the condom attitude and sensation-seeking scales and unrelated psychological constructs such as generalized social desirability, intelligence, and non-health-related locus of control ($r < .12$, non-significant).

Criterion and Predictive Validity

The definitive empirical criterion for the NSHS is its capacity to predict prospective sexual risk events, barrier failure, and laboratory-confirmed biological endpoints. As documented in the foundational baseline publication by Catania et al. (1992) in Science, the instrument demonstrated profound predictive validity across a national probability sample of over 2,600 respondents. Self-reported risk categorizations and barrier negotiation scores predicted documented patterns of condom non-use among populations with multiple partners. Furthermore, in clinical follow-up sub-studies, the psychological scales predicted prospective biological STI diagnoses (including chlamydial infection and gonorrhea), confirming strong criterion-related predictive validity.

Reliability

The reliability of the CAPS National Sexual Health Survey has been systematically confirmed across independent dimensions, encompassing internal consistency of its psychological indices and temporal stability (test-retest reliability) of its sensitive behavioral recall matrices.

Internal Consistency Reliability

The latent psychological scales embedded within the NSHS exhibit high internal consistency, quantified via Cronbach’s coefficient alpha across diverse population sub-samples:

• Condom Attitude Scale: Demonstrates high internal consistency across its full-scale implementation ($lpha = .83$), with sub-dimensions such as Affective Sensory Loss ($lpha = .86$) and Interpersonal Disruption ($lpha = .79$) maintaining robust item-total correlations exceeding .45.
• Perceived Risk / Optimistic Bias Scale: Exhibits an overall internal consistency of $lpha = .78$, confirming cohesive measurement of cognitive vulnerability appraisals.
• Sensation-Seeking Short Form: Achieves a Cronbach’s alpha of $lpha = .81$, demonstrating strong psychometric homogeneity in capturing trait-based disinhibition and novelty preference.
• Machismo / Traditional Masculine Norms Scale: Yields reliability coefficients ranging from $lpha = .77$ to $lpha = .82$ across ethnically stratified subsamples (including English- and Spanish-speaking respondents).

Test-Retest Stability

To assess the temporal consistency of retrospective behavioral recall in telephone interviews, methodological sub-studies conducted test-retest administrations over intervals ranging from two to four weeks. Concordance for categorical behavioral classifications (such as binary lifetime HIV testing status, sexual orientation identification, and lifetime history of STIs) yielded high reliability coefficients, with Cohen’s kappa ($kappa$) values consistently falling between .84 and .94. Continuous behavioral recall metrics—such as the number of vaginal or anal intercourse episodes and specific condom counts in the antecedent 12 months—demonstrated intra-class correlation coefficients (ICC) ranging from .72 to .86, reflecting stable behavioral reporting across time.

Factor Analysis

The dimensional validity of the multi-item psychological batteries embedded within the NSHS has been evaluated through exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) across national probability cohorts.

Exploratory Factor Structure

Initial exploratory factor extractions utilizing principal axis factoring with promax (oblique) and varimax (orthogonal) rotations across the attitudinal items isolated a clear multidimensional architecture. The latent space governing condom attitudes yielded three prominent factors accounting for over 58% of the cumulative variance:

• Factor 1: Sensory and Physical Inhibition (Eigenvalue = 4.12; variance explained = 29.4%). High primary item loadings (.62 to .84) were demonstrated by items assessing diminished sexual pleasure, reduced sensation, interruption of foreplay, and difficulties in sustaining erection during barrier application.
• Factor 2: Interpersonal and Partner Mistrust (Eigenvalue = 2.45; variance explained = 17.5%). Dominated by items reflecting social scripting concerns (.58 to .79), such as the perception that introducing condoms implies promiscuity, infidelity, or accusations of infection.
• Factor 3: Mechanical Inefficacy and Failure Anxiety (Eigenvalue = 1.56; variance explained = 11.1%). Characterized by items loading (.54 to .72) on concerns regarding condom rupture, slippage, and manufacturing inadequacy.

Confirmatory Factor Analysis and Model Fit

Subsequent CFA modeling on cross-validation validation cohorts confirmed the structural integrity of these latent dimensions. First- and second-order structural models demonstrated acceptable to excellent goodness-of-fit indices:

• Comparative Fit Index (CFI) = .952
• Tucker-Lewis Index (TLI) = .941
• Root Mean Square Error of Approximation (RMSEA) = .044 (90% Confidence Interval: [.038, .051])
• Standardized Root Mean Square Residual (SRMR) = .039

Factorial invariance testing further verified that this measurement model exhibits metric and scalar invariance across gender (men vs. women) and broad age categories, establishing that the structural properties of the constructs remain invariant across sociodemographic segments of the U.S. population.

Instrument / Measurement Tool

• Instrument Type: Standardized epidemiological screening protocol, respondent-selection mechanism, and multi-module psychological/behavioral survey interview battery.
• Administration Format: Probability-based Computer-Assisted Telephone Interviewing (CATI) or standardized pencil-and-paper telephone interview administration.
• Target Population: Non-institutionalized adults aged 18 years and older residing in household units within the 48 contiguous United States.
• Household Selection Screener:
  • Household Random Selection Method: Unclustered “next-birthday” adult respondent selection technique.
  • Eligibility Verification: Confirmation of residence, residential telephone line, and age qualification (birthdate verified as 18 years of age or older at the time of screening).
  • Callback Protocol: Structured tracking matrix recording respondent first name, designated callback date and time, telephone verification, and interviewer identification.
  • Informed Consent Module: Standardized verbal disclosure detailing research intent, institutional sponsor (University of California), public health utility, strict confidentiality assurances, right of selective item refusal, and availability of an institutional toll-free verification hotline.
• Survey Item Count: The instrument comprises a modular architecture including a multi-item standardized screening script, followed by the comprehensive survey battery containing approximately 150 to 250 modular questions depending on partner-skip branching logic (tracking up to 10 distinct sexual partners over the antecedent 12 months).
• Response Formats: Categorical selection indicators, numeric branch skips, structured callback schedules, exact calendar birthdate entries, and multiple standardized response scales (Likert-type scales, frequency ratings, and dichotomous Yes/No options).
• Scoring and Quantification Rules:
  • Screener Branching: Strict skip routing dictates interviewer progression based on household adult availability (e.g., informant selection skips directly to informed consent; selection of another adult routes to respondent availability and scheduling).
  • Attitudinal and Psychological Indices: Summed or standardized mean composite scores across latent scales (e.g., Sensation-Seeking, Condom Attitudes), with designated reverse-coded items ensuring unidirectional alignment with theoretical risk constructs.
  • Behavioral Dyadic Matrices: Numerical summation of total sexual encounters, discrete counts of protected and unprotected penetrative events, and derived proportions of consistent barrier utilization calculated per partner and aggregated across partners.

Permissions & Fee and Test Year

The CAPS National Sexual Health Survey was developed and first executed at a national scale in 1990–1992, with initial landmark baseline findings published in Science in 1992. The instrument was funded through major federal research grants awarded by the National Institutes of Health (NIH), including the National Institute of Mental Health (NIMH) and the National Institute on Alcohol Abuse and Alcoholism (NIAAA), to the Center for AIDS Prevention Studies (CAPS) at the University of California, San Francisco.

As a federally funded public health surveillance tool, the NSHS screening protocols and questionnaire modules are in the public domain for non-commercial research, epidemiological surveillance, and educational evaluation purposes. No licensing fees or royalties are required to adapt or utilize the screening instrument or measurement scales. However, professional psychometric and epidemiological standards necessitate appropriate scholarly citation of the original CAPS design team and the foundational 1992 publication in any subsequent adaptations, empirical studies, or publications.

References

• Catania, J. A., Coates, T. J., & Kegeles, S. (1994). A test of the AIDS Risk Reduction Model: Psychological predictors of condom use among gay men. Health Psychology, 13(6), 548–555. https://doi.org/10.1037/0278-6133.13.6.548
• Catania, J. A., Coates, T. J., Stall, R., Turner, H. A., Peterson, J., Hearst, N., Dolcini, M. M., Hudes, E., Gagnon, J., Wiley, J., & Groves, R. (1992). Prevalence of AIDS-related risk factors and condom use in the United States. Science, 258(5085), 1101–1106. https://doi.org/10.1126/science.1439823
• Catania, J. A., Gibson, D. R., Chitwood, D. D., & Coates, T. J. (1990). Methodological problems in AIDS behavioral research: Influences on measurement error and participation bias in studies of sexual behavior. Psychological Bulletin, 108(3), 339–362. https://doi.org/10.1037/0033-2909.108.3.339
• Catania, J. A., Kegeles, S. M., & Coates, T. J. (1990). Towards an understanding of risk behavior: An AIDS Risk Reduction Model (ARRM). Health Education Quarterly, 17(1), 53–72. https://doi.org/10.1177/109019819001700107
• Dolcini, M. M., Catania, J. A., Coates, T. J., Stall, R., Hudes, E. S., Gagnon, J. H., & Pollack, L. M. (1993). Demographic characteristics of heterosexuals with multiple partners: The National AIDS Behavioral Surveys. Family Planning Perspectives, 25(5), 208–214. https://doi.org/10.2307/2135898
• Groves, R. M., & Lyberg, L. (1988). An overview of nonresponse issues in telephone surveys. In R. M. Groves, P. P. Biemer, L. E. Lyberg, J. T. Massey, W. L. Nicholls, & J. Waksberg (Eds.), Telephone Survey Methodology (pp. 191–212). John Wiley & Sons. https://doi.org/10.1002/9781118150498.ch12
• Salmon, C. T., & Nichols, J. S. (1983). The next-birthday method of respondent selection. Public Opinion Quarterly, 47(2), 270–276. https://doi.org/10.1086/268784
• Zuckerman, M. (1994). Behavioral expressions and biosocial bases of sensation seeking. Cambridge University Press. https://doi.org/10.1017/CBO9780511752957

Items of the Scale